/**
* Enhanced Prompt Templates
* Defines the four sophisticated prompt templates that replace static prompts
* with dynamic, context-aware generation capabilities
*/
import { PromptTemplate } from '../types/enhanced-prompts.js';
/**
* Template 1: analyze-dataproc-query (Knowledge-Driven)
* Analyzes queries with knowledge base examples and optimization insights
*/
export const analyzeDataprocQueryTemplate: PromptTemplate = {
id: 'analyze-dataproc-query',
category: 'knowledge-driven',
description: 'Analyze Hive/Spark queries with knowledge base examples and optimization insights',
template: `Analyze this {{queryType || 'SQL'}} query for optimization and best practices:
**Query to Analyze:**
\`\`\`sql
{{query}}
\`\`\`
**Context:**
- Target Cluster: {{clusterName || 'any'}}
- Optimization Level: {{optimizationLevel || 'basic'}}
- Query Type: {{queryType || 'auto-detect'}}
- Project: {{projectId || 'not specified'}}
- Region: {{region || 'not specified'}}
**Similar Query Examples from Knowledge Base:**
{{qdrant_query("similar queries " + queryType, "limit:3 type:job")}}
**Performance Insights from Cluster Data:**
{{qdrant_query("query optimization " + queryType + " performance", "limit:2 type:cluster")}}
**Historical Query Patterns:**
{{qdrant_query("successful " + queryType + " patterns", "limit:2 type:job")}}
**Analysis Requirements:**
Please provide a comprehensive analysis including:
1. **Query Performance Analysis**
- Identify potential bottlenecks and performance issues
- Analyze query complexity and resource requirements
- Evaluate join strategies and data access patterns
2. **Optimization Suggestions**
- Specific recommendations based on similar successful queries
- Indexing and partitioning strategies
- Query rewriting opportunities for better performance
3. **Best Practices Recommendations**
- Industry best practices for {{queryType}} queries
- Dataproc-specific optimizations
- Resource allocation recommendations
4. **Potential Issues and Warnings**
- Common pitfalls and anti-patterns detected
- Resource consumption warnings
- Compatibility considerations
5. **Alternative Query Approaches**
- Different strategies to achieve the same result
- Trade-offs between approaches
- Performance vs. complexity considerations`,
parameters: [
{
name: 'query',
type: 'string',
required: true,
source: 'tool',
description: 'SQL/HiveQL query to analyze',
},
{
name: 'queryType',
type: 'string',
required: false,
source: 'tool',
description: 'Type of query (hive, spark, presto, auto)',
validation: {
enum: ['hive', 'spark', 'presto', 'auto'],
},
},
{
name: 'clusterName',
type: 'string',
required: false,
source: 'profile',
description: 'Target cluster name for optimization context',
},
{
name: 'optimizationLevel',
type: 'string',
required: false,
source: 'tool',
description: 'Level of optimization analysis',
validation: {
enum: ['basic', 'advanced', 'expert'],
},
defaultValue: 'basic',
},
{
name: 'projectId',
type: 'string',
required: false,
source: 'gcp',
description: 'GCP project ID for context',
},
{
name: 'region',
type: 'string',
required: false,
source: 'gcp',
description: 'GCP region for context',
},
],
knowledgeQueries: [
{
type: 'semantic',
query: 'similar queries {{queryType}}',
filters: { type: 'job' },
limit: 3,
minConfidence: 0.7,
},
{
type: 'semantic',
query: 'query optimization {{queryType}} performance',
filters: { type: 'cluster' },
limit: 2,
minConfidence: 0.6,
},
{
type: 'semantic',
query: 'successful {{queryType}} patterns',
filters: { type: 'job' },
limit: 2,
minConfidence: 0.7,
},
],
dynamicFunctions: [
{
name: 'qdrant_query',
args: ['similar queries {{queryType}}', 'limit:3 type:job'],
cacheKey: 'similar_queries_{{queryType}}',
ttl: 1800000, // 30 minutes
},
{
name: 'qdrant_query',
args: ['query optimization {{queryType}} performance', 'limit:2 type:cluster'],
cacheKey: 'optimization_{{queryType}}',
ttl: 3600000, // 1 hour
},
{
name: 'qdrant_query',
args: ['successful {{queryType}} patterns', 'limit:2 type:job'],
cacheKey: 'patterns_{{queryType}}',
ttl: 1800000, // 30 minutes
},
],
};
/**
* Template 2: design-dataproc-cluster (Resource-Integrated)
* Generates cluster configurations with profile integration and regional best practices
*/
export const designDataprocClusterTemplate: PromptTemplate = {
id: 'design-dataproc-cluster',
category: 'resource-integrated',
description: 'Generate cluster configuration recommendations with profile integration',
template: `Design an optimal Dataproc cluster configuration for the following requirements:
**Workload Requirements:**
- **Type**: {{workloadType}}
- **Data Size**: {{dataSize}}
- **Budget**: {{budget}}
- **Region**: {{region || 'flexible'}}
- **Additional Requirements**: {{requirements || 'none specified'}}
**Profile-Based Configuration Templates:**
{{profile_config(profileId, "cluster-templates")}}
**Similar Cluster Configurations from Knowledge Base:**
{{qdrant_query("workload " + workloadType + " " + dataSize + " cluster", "limit:3 type:cluster")}}
**Regional Best Practices and Optimizations:**
{{qdrant_query("region " + region + " cluster configuration best practices", "limit:2")}}
**Cost Optimization Patterns:**
{{qdrant_query(budget + " budget cluster optimization", "limit:2 type:cluster")}}
**Design Requirements:**
Please provide a comprehensive cluster design including:
1. **Recommended Cluster Configuration**
- Master node specifications (machine type, disk size, count)
- Worker node specifications (machine type, disk size, count)
- Preemptible worker configuration if applicable
- Network and security group settings
2. **Software Components and Versions**
- Recommended Dataproc image version for {{workloadType}}
- Optional components based on workload requirements
- Initialization scripts and custom configurations
3. **Networking and Security Recommendations**
- VPC and subnet configuration for {{region}}
- Firewall rules and security best practices
- Service account and IAM permissions
- Encryption settings (at-rest and in-transit)
4. **Cost Optimization Strategies**
- Strategies specific to {{budget}} budget constraints
- Preemptible instance recommendations
- Auto-scaling policies for cost efficiency
- Resource scheduling and lifecycle management
5. **Performance Tuning Suggestions**
- Configuration optimizations for {{dataSize}} data volumes
- Memory and CPU allocation strategies
- Storage optimization (SSD vs HDD, local vs persistent)
- Network performance considerations
6. **Complete YAML Configuration File**
- Ready-to-use cluster configuration
- Environment-specific parameter substitution
- Validation and deployment instructions
7. **Scaling and Monitoring Recommendations**
- Auto-scaling policies and thresholds
- Monitoring and alerting setup
- Performance metrics to track
- Maintenance and upgrade strategies`,
parameters: [
{
name: 'workloadType',
type: 'string',
required: true,
source: 'tool',
description: 'Type of workload the cluster will handle',
validation: {
enum: ['analytics', 'ml', 'streaming', 'batch', 'mixed'],
},
},
{
name: 'dataSize',
type: 'string',
required: true,
source: 'tool',
description: 'Expected data size to process',
validation: {
enum: ['small', 'medium', 'large', 'xlarge'],
},
},
{
name: 'budget',
type: 'string',
required: true,
source: 'tool',
description: 'Budget constraints for the cluster',
validation: {
enum: ['low', 'medium', 'high', 'unlimited'],
},
},
{
name: 'region',
type: 'string',
required: false,
source: 'gcp',
description: 'Target GCP region for deployment',
},
{
name: 'requirements',
type: 'string',
required: false,
source: 'tool',
description: 'Additional specific requirements or constraints',
},
{
name: 'profileId',
type: 'string',
required: false,
source: 'profile',
description: 'Profile ID for configuration templates',
},
],
resourceAccess: [
{
uri: 'dataproc://profile/{{category}}/{{profileId}}/config',
purpose: 'cluster-templates',
cacheKey: 'profile_{{profileId}}_templates',
ttl: 3600000, // 1 hour
},
],
knowledgeQueries: [
{
type: 'semantic',
query: 'workload {{workloadType}} {{dataSize}} cluster',
filters: { type: 'cluster' },
limit: 3,
minConfidence: 0.7,
},
{
type: 'semantic',
query: 'region {{region}} cluster configuration best practices',
limit: 2,
minConfidence: 0.6,
},
{
type: 'semantic',
query: '{{budget}} budget cluster optimization',
filters: { type: 'cluster' },
limit: 2,
minConfidence: 0.6,
},
],
dynamicFunctions: [
{
name: 'profile_config',
args: ['{{profileId}}', 'cluster-templates'],
cacheKey: 'profile_config_{{profileId}}',
ttl: 3600000, // 1 hour
},
{
name: 'qdrant_query',
args: ['workload {{workloadType}} {{dataSize}} cluster', 'limit:3 type:cluster'],
cacheKey: 'workload_{{workloadType}}_{{dataSize}}',
ttl: 1800000, // 30 minutes
},
],
};
/**
* Template 3: troubleshoot-dataproc-issue (Fully-Integrated)
* Comprehensive troubleshooting with all system integrations
*/
export const troubleshootDataprocIssueTemplate: PromptTemplate = {
id: 'troubleshoot-dataproc-issue',
category: 'fully-integrated',
description: 'Comprehensive troubleshooting with semantic error matching and job analysis',
template: `Troubleshoot this Dataproc issue with comprehensive analysis:
**Issue Details:**
- **Type**: {{issueType}}
- **Error Message**: {{errorMessage || 'not provided'}}
- **Job ID**: {{jobId || 'not applicable'}}
- **Cluster**: {{clusterName || 'not specified'}}
- **Timeline**: {{timeline || 'not specified'}}
- **Context**: {{context || 'not provided'}}
- **Project**: {{projectId || 'not specified'}}
- **Region**: {{region || 'not specified'}}
**Job Output and Error Analysis:**
{{job_output(jobId, "error-analysis")}}
**Similar Error Patterns from Knowledge Base:**
{{qdrant_query("error " + issueType + " " + errorMessage, "limit:5 type:error")}}
**Cluster Configuration Analysis:**
{{qdrant_query("cluster " + clusterName + " configuration issues", "limit:3 type:cluster")}}
**Historical Resolution Patterns:**
{{qdrant_query("resolved " + issueType + " troubleshooting solutions", "limit:3")}}
**Related Job Failures:**
{{qdrant_query("job failure " + issueType + " " + clusterName, "limit:2 type:job")}}
**Troubleshooting Analysis:**
Please provide a systematic troubleshooting approach:
1. **Root Cause Analysis**
- Primary cause identification based on error patterns
- Secondary contributing factors
- System state analysis at time of failure
- Resource utilization assessment
2. **Step-by-Step Troubleshooting Guide**
- Immediate diagnostic steps to take
- Data collection and log analysis procedures
- Progressive isolation techniques
- Verification checkpoints
3. **Diagnostic Commands and Tools**
- Specific gcloud commands for investigation
- Log queries and monitoring dashboards
- Performance analysis tools
- Network and connectivity tests
4. **Resolution Strategies**
- Primary resolution approach based on similar cases
- Alternative solutions if primary approach fails
- Rollback procedures if needed
- Testing and validation steps
5. **Prevention Recommendations**
- Configuration changes to prevent recurrence
- Monitoring and alerting improvements
- Best practices implementation
- Proactive maintenance procedures
6. **Escalation Path**
- When to involve Google Cloud Support
- Information to gather before escalation
- Internal team escalation procedures
- Emergency response protocols`,
parameters: [
{
name: 'issueType',
type: 'string',
required: true,
source: 'tool',
description: 'Type of issue being experienced',
validation: {
enum: ['job-failure', 'cluster-startup', 'performance', 'connectivity', 'other'],
},
},
{
name: 'errorMessage',
type: 'string',
required: false,
source: 'tool',
description: 'Specific error message or symptoms',
},
{
name: 'jobId',
type: 'string',
required: false,
source: 'tool',
description: 'Job ID if the issue is related to a specific job',
},
{
name: 'clusterName',
type: 'string',
required: false,
source: 'tool',
description: 'Name of the affected cluster',
},
{
name: 'timeline',
type: 'string',
required: false,
source: 'tool',
description: 'When the issue started or was first noticed',
},
{
name: 'context',
type: 'string',
required: false,
source: 'tool',
description: 'Additional context about what was being attempted',
},
{
name: 'projectId',
type: 'string',
required: false,
source: 'gcp',
description: 'GCP project ID for context',
},
{
name: 'region',
type: 'string',
required: false,
source: 'gcp',
description: 'GCP region for context',
},
],
dynamicFunctions: [
{
name: 'job_output',
args: ['{{jobId}}', 'error-analysis'],
cacheKey: 'job_output_{{jobId}}',
ttl: 600000, // 10 minutes
},
{
name: 'qdrant_query',
args: ['error {{issueType}} {{errorMessage}}', 'limit:5 type:error'],
cacheKey: 'error_{{issueType}}_patterns',
ttl: 1800000, // 30 minutes
},
{
name: 'qdrant_query',
args: ['cluster {{clusterName}} configuration issues', 'limit:3 type:cluster'],
cacheKey: 'cluster_{{clusterName}}_issues',
ttl: 900000, // 15 minutes
},
{
name: 'qdrant_query',
args: ['resolved {{issueType}} troubleshooting solutions', 'limit:3'],
cacheKey: 'resolved_{{issueType}}',
ttl: 3600000, // 1 hour
},
],
knowledgeQueries: [
{
type: 'semantic',
query: 'error {{issueType}} {{errorMessage}}',
filters: { type: 'error' },
limit: 5,
minConfidence: 0.6,
},
{
type: 'semantic',
query: 'cluster {{clusterName}} configuration issues',
filters: { type: 'cluster' },
limit: 3,
minConfidence: 0.7,
},
{
type: 'semantic',
query: 'resolved {{issueType}} troubleshooting solutions',
limit: 3,
minConfidence: 0.7,
},
],
};
/**
* Template 4: generate-dataproc-query (Template-Enhanced)
* Generates optimized queries with template-driven optimization
*/
export const generateDataprocQueryTemplate: PromptTemplate = {
id: 'generate-dataproc-query',
category: 'template-enhanced',
description: 'Generate optimized queries with template-driven optimization and examples',
template: `Generate an optimized {{dialect}} query for the following requirements:
**Requirements:**
- **Purpose**: {{queryPurpose}}
- **Tables**: {{tableNames || 'to be determined'}}
- **Query Type**: {{queryType}}
- **Dialect**: {{dialect}}
- **Performance Level**: {{performanceLevel}}
- **Target Cluster**: {{clusterName || 'any'}}
- **Project**: {{projectId || 'not specified'}}
- **Region**: {{region || 'not specified'}}
**Cluster-Specific Optimization Hints:**
{{cluster_status(clusterName, "optimization-hints")}}
**Query Pattern Examples from Knowledge Base:**
{{qdrant_query("successful " + queryType + " " + dialect + " queries", "limit:3 type:job")}}
**Performance Best Practices:**
{{qdrant_query(dialect + " " + performanceLevel + " optimization patterns", "limit:2")}}
**Similar Query Implementations:**
{{qdrant_query("query " + queryPurpose + " " + dialect, "limit:2 type:job")}}
**Generation Requirements:**
Please provide a complete query solution including:
1. **Complete, Runnable Query**
- Fully functional {{dialect}} query that addresses the requirements
- Proper syntax and formatting for {{dialect}}
- Comments explaining key sections and logic
- Parameter placeholders where appropriate
2. **Query Logic and Approach Explanation**
- Step-by-step breakdown of the query logic
- Reasoning behind chosen approach
- Data flow and transformation explanation
- Join strategies and filtering logic
3. **Performance Optimizations**
- Specific optimizations for {{performanceLevel}} performance level
- Indexing recommendations for better performance
- Partitioning strategies if applicable
- Memory and resource optimization techniques
4. **Alternative Approaches**
- Different strategies to achieve the same result
- Trade-offs between different approaches
- When to use each alternative approach
- Performance implications of each option
5. **Execution Considerations**
- Memory requirements and allocation strategies
- Parallelism and concurrency considerations
- Resource usage patterns and optimization
- Estimated execution time and resource consumption
6. **Monitoring and Validation Recommendations**
- Key metrics to monitor during execution
- Performance benchmarks and thresholds
- Data quality validation checks
- Error handling and recovery strategies`,
parameters: [
{
name: 'queryPurpose',
type: 'string',
required: true,
source: 'tool',
description: 'What the query is intended to accomplish',
},
{
name: 'tableNames',
type: 'string',
required: false,
source: 'tool',
description: 'Comma-separated list of table names to work with',
},
{
name: 'queryType',
type: 'string',
required: true,
source: 'tool',
description: 'Type of query operation',
validation: {
enum: ['select', 'insert', 'update', 'create', 'analyze'],
},
},
{
name: 'dialect',
type: 'string',
required: true,
source: 'tool',
description: 'SQL dialect to use',
validation: {
enum: ['hive', 'spark-sql', 'presto'],
},
},
{
name: 'performanceLevel',
type: 'string',
required: true,
source: 'tool',
description: 'Target performance level',
validation: {
enum: ['standard', 'optimized', 'high-performance'],
},
},
{
name: 'clusterName',
type: 'string',
required: false,
source: 'profile',
description: 'Target cluster for optimization context',
},
{
name: 'projectId',
type: 'string',
required: false,
source: 'gcp',
description: 'GCP project ID for context',
},
{
name: 'region',
type: 'string',
required: false,
source: 'gcp',
description: 'GCP region for context',
},
],
templateResolution: {
enableParameterInjection: true,
enableDynamicFunctions: true,
cacheResults: true,
timeoutMs: 15000,
},
dynamicFunctions: [
{
name: 'cluster_status',
args: ['{{clusterName}}', 'optimization-hints'],
cacheKey: 'cluster_status_{{clusterName}}',
ttl: 900000, // 15 minutes
},
{
name: 'qdrant_query',
args: ['successful {{queryType}} {{dialect}} queries', 'limit:3 type:job'],
cacheKey: 'successful_{{queryType}}_{{dialect}}',
ttl: 1800000, // 30 minutes
},
{
name: 'qdrant_query',
args: ['{{dialect}} {{performanceLevel}} optimization patterns', 'limit:2'],
cacheKey: 'optimization_{{dialect}}_{{performanceLevel}}',
ttl: 3600000, // 1 hour
},
],
knowledgeQueries: [
{
type: 'semantic',
query: 'successful {{queryType}} {{dialect}} queries',
filters: { type: 'job' },
limit: 3,
minConfidence: 0.7,
},
{
type: 'semantic',
query: '{{dialect}} {{performanceLevel}} optimization patterns',
limit: 2,
minConfidence: 0.6,
},
{
type: 'semantic',
query: 'query {{queryPurpose}} {{dialect}}',
filters: { type: 'job' },
limit: 2,
minConfidence: 0.6,
},
],
};
/**
* All enhanced prompt templates
*/
export const ENHANCED_PROMPT_TEMPLATES: PromptTemplate[] = [
analyzeDataprocQueryTemplate,
designDataprocClusterTemplate,
troubleshootDataprocIssueTemplate,
generateDataprocQueryTemplate,
];
/**
* Template registry for easy access
*/
export const TEMPLATE_REGISTRY = new Map<string, PromptTemplate>(
ENHANCED_PROMPT_TEMPLATES.map((template) => [template.id, template])
);
